{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/optimizing-deep-neural-network-architecture-a","title":"Optimizing Deep Neural Network Architecture: A Tabu Search Based Approach","arxiv_id":"1808.05979","date":"2018-08-17","proceeding":null,"authors":["Tarun Kumar Gupta","Khalid Raza"],"abstract":"The performance of Feedforward neural network (FNN) fully de-pends upon the\nselection of architecture and training algorithm. FNN architecture can be\ntweaked using several parameters, such as the number of hidden layers, number\nof hidden neurons at each hidden layer and number of connections between\nlayers. There may be exponential combinations for these architectural\nattributes which may be unmanageable manually, so it requires an algorithm\nwhich can automatically design an optimal architecture with high generalization\nability. Numerous optimization algorithms have been utilized for FNN\narchitecture determination. This paper proposes a new methodology which can\nwork on the estimation of hidden layers and their respective neurons for FNN.\nThis work combines the advantages of Tabu search (TS) and Gradient descent with\nmomentum backpropagation (GDM) training algorithm to demonstrate how Tabu\nsearch can automatically select the best architecture from the populated\narchitectures based on minimum testing error criteria. The proposed approach\nhas been tested on four classification benchmark dataset of different size.","url_abs":"http://arxiv.org/abs/1808.05979v1","url_pdf":"http://arxiv.org/pdf/1808.05979v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"optimizing-deep-neural-network-architecture-a","repo_url":"https://github.com/marzekan/Digit-classifier","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}